Bariatric surgery for diabetic teens — in their best interests?

Research that enables pediatricians to provide the best care for their young patients is something many wonderful doctors truly care about, and certainly parents do, too. That’s why it is unimaginable this article appeared in a medical journal. There is simply no credible justification for its publication.

This article in the January issue of Pediatrics, the journal of the American Academy of Pediatrics, wasn’t a clinical trial. In fact, it violated basic principles of medical research and offered no information that any credible healthcare professional would actually use in the care of patients. Its unsupportable conclusions were further sensationalized beyond recognition in a press release, with claims not remotely supported by the findings or the methodology. Yet, it was devoured by gullible journalists and resulted in 331 dramatic news stories in a single day, with amazing promises that, no doubt, brought countless new customers to the lead medical center.

Media script

The headlines have been saying that “Bariatric surgery reverses type 2 diabetes in teens.” The media stories have followed the press release issued on Monday by Cincinnati Children’s Hospital Medical Center. Its tone was similarly breathless, reporting “dramatic, often immediate remission,” using the word dramatic four times, remission five times, and diabetes-free twice, along with breakthrough, impressive, and significant. Through a testimonial, claims were made that the surgery resulted in the loss of one-third of body weight and that “remaining diabetes free is well worth” having the surgery.

“Until now,” the news release said, “little information was available for families considering surgical weight loss for adolescents.” But according to Dr. Thomas Inge, M.D., Ph.D., lead author and surgical director of the teen bariatric program at Cincinnati Children’s: “The results have been quite dramatic and to our knowledge, there are no other anti-diabetic therapies that result in more effective and long-term control than that seen with bariatric surgery.”

“Dr. Inge and his colleagues agree that the numerous benefits of such procedures will likely outweigh the risks for qualified surgical candidates,” it said.

This press release included an obesity epidemic fact sheet from Ethicon-Endo Surgery. Accompanying the print press release, was a video news release by Dr. Inge to explain this “breakthrough study.” In it, he also described the “extraordinary results” of a “complete remission” of diabetes and “dramatic reduction” in risks for “heart disease, renal failure and blindness” with “enormous implications for their future health.” While there can be complications, these were quickly brushed aside as he said in the same breath, “the real story is that the risks of diabetes may be completely diverted.”

According to the authors, bariatric surgeries are critically needed because of rising rates of obesity and type 2 diabetes in children and adolescents. “We previously found a ten-fold increase in the incidence of adolescent type 2 diabetes in Cincinnati from 1982 and 1994,” they wrote. [Yes, this is the same Cincinnati Children’s Hospital that was the source for the widely repeated fallacy of an epidemic of type 2 diabetes among children and teens.]

But the bariatric study being conducted by these authors has barely begun and is years away from completion. How can long-term effectiveness, extraordinary preventive abilities, and safety have already been shown?

They can’t and haven’t.

The authors of this study are with the Teen-Longitudinal Assessment of Bariatric Surgery (Teen-LABS) consortium, the five centers with bariatric surgery units awarded a $3.9 million NIH grant in June 2006. Cincinnati Children’s is the coordinating center. Teen-LABS is to be the first-ever study of bariatric surgery on 200 teens to help “determine if it is an appropriate treatment option for extremely overweight teens... and help them remain at a healthy weight over the long-term.” This observational study began enrolling teens in March 2007 and is slated to end in July 2011.

Research or marketing?

Also linked to the Cincinnati Children’s press release page is the brochure for its teen bariatric surgical unit, as well as a Teen-LABS in the News section.

How many research studies, purportedly designed to assess the safety and efficacy of a new procedure, have advance publicity featuring before-and-after success stories on Oprah, NBC’s Today Show and various consumer publications? Researchers who set out to objectively answer a scientific question do not use press releases or go on television shows to report the findings years before their study is even completed.

As we’ve examined, however, the focus of Teen-LAPS is less on investigating long-term complications and effects on health (which have previously been identified as concerns by these same investigators), and more about identifying demographic and lifestyle information. But no one even has to read the study protocol, to realize that this study isn’t doing any intensive or long-term clinical follow-up of medical, nutritional and neurological complications for these young people. That’s evident just looking at the study brochure for teens and the specific things they are told they will be asked to do if they participate:

Incredibly, this study in Pediatrics wasn’t even reporting preliminary data from Teen-LABS. What was it reporting?

A study that isn’t a study

This article was a retrospective chart review of eleven teens who had gastric bypass at the various Teen-LAPS bariatric centers between November 2002 and 2004. Select health indices of these teens at one year were compared to those in the electronic medical records of a control group of 53 teens with type 2 diabetes who had been seen at Cincinnati Children’s Hospital between 2002 and April 2005. Case reports on a hand-selected group of patients are not the types of studies that doctors can use to evaluate the effectiveness of an intervention and, most importantly, use to weigh the risks and benefits. These are akin to anecdotal evidence.

The first obvious flaw that negates this as a “fair test” of anything is selection bias. Selection bias is like cherry picking. The children chosen to be reported on, after the fact, didn’t include all of the kids in a population who had surgery, nor were they randomly selected from the patient rosters. They were hand picked. Randomization is critical to help ensure that we’re not just hearing about the rosiest outcomes, and that the study group reflects the typical patient.

These eleven young people also represent an infinitesimally tiny percentage — less than 0.4% — of the bariatric surgeries that have been performed on teens. An analysis of national trends in bariatric surgeries for teens was published in the March 2007 issue of Archives of Pediatric and Adolescent Medicine. It was conducted by Dr. Inge, along with colleagues at Robert Wood Johnson Medical School at the University of Medicine & Dentistry of New Jersey. They estimated that 2,744 bariatric surgeries had been performed on adolescents in the United Stated between 1996 and 2003, with 771 in 2003 alone. More than 70 teens had had bariatric surgery just at Cincinnati Children’s since 2001, said Dr. Inge in an accompanying press release two years ago.

Yet, only eleven were chosen.

We know nothing about what happened to the thousands of other teens who had bariatric surgery, so no one can possibly conclude that the surgery is a safe medical intervention. How many of those others died or have been left with innumerable complications? How have their physical and psychological health, growth and development been affected? Eleven anecdotal reports, just like testimonials, do not allow us to weigh the risks.

When the control group being used for comparison is also hand picked, it increases risks for stacking the deck. Girls made up about 60% of both groups in this report, but otherwise, the control group bore little resemblance to the 11 bariatric surgery patients. They were 2 ½ years younger — average age nearly 15.5 years at the start of the year, compared to 18 years of age in the surgical group. They weighed an average of 105.6 pounds less (with only 8 kids with BMIs in the range of the bariatric group). Fasting blood sugars, insulin levels or HOMA-2 scores (estimates of insulin resistance, although normal values for adolescents have never been published) were not reported for the control group, but they did have 7-26% higher HbA1c levels (surrogate measures of glucose levels). It is well-known that a normal part of human development is the transient insulin resistance that occurs during puberty, moreso in girls and Blacks, which drops down to nearly prepubertal levels by early adulthood. There were no Blacks among the bariatric patient group and race/ethnicity was not reported for the control group. HbA1c levels also vary according to duration of diabetes, being highest during the first two years then dropping. No information was given about the duration of diabetes among either selected group, although the higher levels seen in the younger control group might also have been weighted by duration of their disease.

In other words, no credible comparisons between these very different groups, with key variables uncontrolled, can be made.

This study claimed that in all but one of the teens, their type 2 diabetes had been cured almost before leaving the hospital. This was based on discontinued glycemic medication use during the first 12 months post-op. One explanation why weight loss didn’t eliminate the need for medication in that young man was that his diabetes was more advanced than the other adolescents, said Dr. Inge.

The authors admitted that not only did they have limited lab results on the kids, but also that there was no standardization of laboratory assays for any of the lab tests reported in this study. At the end of the one-year study period, HbA1c levels were only available for 5 of the 11 teens. Yet, incredibly, the authors stated that they had “for the first time, to our knowledge, detailed the clinical and metabolic changes that might be expected for adolescents with type 2 diabetes who undergo RYGB.”

No pediatrician would actually consider lab results on five kids to be detailed clinical information or to give an accurate indication of what can be expected for all patients referred for bariatric surgery.

More important, a drop in blood sugar does not mean the underlying disease pathology of diabetes has been cured or reversed. Blood sugars are a symptom, a surrogate endpoint, not the disease of diabetes itself. [This popularly misunderstood distinction was covered here.] Transient drops in blood glucose levels and other health indices during the short-term clinical trials done to date have reflected the degree of caloric restriction and weight loss, not actual glycemic improvement, or shown to be sustained, as Dr. Rubino and Dr. Jacques Marescaux, M.D., FRCS at the University Louis Pasteur in France explained. Claims that bariatric surgery results in a remission of type 2 diabetes are not supported scientifically, they said, and no study has been designed to specifically test if surgery is an effective treatment and leads to better long-term health outcomes for people with type 2 diabetes, and with benefits that outweigh the risks. As they explained:

[A] small, uncontrolled case-series type of study is not the proper instrument to demonstrate a direct effect of surgery on type 2 diabetes as there are several possible reasons that could justify improved glycemia after a bariatric operation. For instance, since patients undergoing Roux-en-Y gastric bypass or biliopancreatic diversion eat small, rather fluid and low-caloric meals in the early postoperative period, it is admittedly impracticable to rule out that the rapid normalization of plasma glucose and improved insulin resistance after these surgeries be simply the effect of decreased caloric intake.

Most disturbing in the claims swirling around this tiny report have been the impressions given that the surgery is safe and will reduce premature deaths from heart disease and other diabetes-related health problems. Besides not reporting any hard clinical outcomes, this study was too short to credibly suggest long-term health benefits — and especially, to investigate risks for long-term complications.

Yet, the authors overstated the conclusions even possible from their case report, writing:

We also demonstrate that a small group of adolescents with T2DM can undergo a major operation safely. These data may prove useful for clinical decision making… The lack of any major medical or surgical complications suggests that the risk/benefit ratio for RYGB in adolescents with T2DM is favorable.

In addition, the magnitude of the improvement in T2DM disease status, as measured by HbA1c and medication usage, may well be superior with surgery.

CONCLUSIONS. Our observations provide evidence that bariatric surgery reverses or significantly improves T2DM in adolescents over a 1-year period, further supporting the role of surgery outlined in recent [AAP] treatment recommendations.

Most readers likely missed the qualifiers can, may, and suggests. Their later disclaimer — “However, the long-term safety and efficacy of bariatric surgery in adolescents remains to be firmly established.” — also had little effect on moderating the sensational news coverage.

Missing balance

Rather than another paper reporting apparent favorable results seen during the initial honeymoon period of weight loss, why not publish a much more valuable follow-up report on how earlier gastric bypass patients are doing six to ten years later? Or even how these 11 patients are doing today? Dr. Inge and colleagues at Cincinnati Children’s have previously published articles about serious complications and the desperate need for long-term data on bariatric surgeries on teens.

As covered in more detail here, in their 2007 review of bariatric procedures in teens, they reported “profound” complications seen among procedures done on adolescents through the mid-1990s. The newest and current variations of gastric bypass surgeries, RYGB, has limited evidence in teens, they said. The largest retrospective study in teens with the longest follow-up to date, they wrote, was on 33 adolescents (average age 16) who had RYGB and variations. The surgeons reported that nine teens had major complications within 30 days of surgery, one died one year later and another at six years. All were still obese at every year measured after surgery. More importantly, they noted, all but six were lost in follow-up, with no data available on their status by 14 years.

Three other centers performing laparoscopic gastric bypass on a total of 41 teens since 2005 reported that 39% had complications, two of which had long-term consequences, including a death. While gastric banding “seems to be a technically safer operation with lower mortality risks than other procedures,” they said, “systematically collected information about both efficacy and potentially worrisome complications (5-10 years) later...for adolescents who have perhaps five or six decades to live with the device” make long-term information about efficacy needed. Despite claims that teens experience fewer post-op complications, recent analyses found they were similar to adults, they said.

Nutritional deficiencies are increased in both restrictive and malabsorption procedures (gastric bypass), they cautioned, and the resulting problems were not insignificant. Peripheral neuropathy is one of the most common vitamin deficiency manifestations, occurring in up to 16% of patients. They described their own 2004 study reporting on three teen girls they were following who experienced neurological complications from thiamine deficiency. The girls, 14 to 17 years of age, developed a range of neurological symptoms, including hearing loss, dizziness, numbness and pain in extremities; vomiting; weakness and loss of strength, and inability to walk, just 4 to 6 months after their surgeries.

Dr. Inge and colleagues went on to caution about increased risks for iron deficiency anemia seen after gastric bypass, especially in menstruating young women, despite oral iron supplementation. They expressed concerns about osteoporosis and osteopenia surfacing among patients a decade out from surgery; although they reported that decreased bone mineral density is evident at 9 months, despite normal parathyroid hormone, calcium and vitamin D levels. “Long term data on these adolescents are not yet available,” they cautioned.

But healthcare professionals and the public heard nothing about these concerns or other potentially serious complications being seen among bariatric patients. Instead, we got a case report on eleven young people who had surgery 5 to 7 years ago, giving limited results after only their first 12 months post-op. It was reported as being a “breakthrough study” showing dramatic results that offer “an optimistic outlook for their future.”

And published in a medical journal.

And not one of those more than 300 media stories headlined with: “New report cautions that bariatric surgeries have not yet been shown to have long-term safety or effectiveness in teens.”

That’s the difference between medical information and marketing.

© 2009 Sandy Szwarc